Background
Carrageenan, a high-molecular weight saccharide, is used widely as a thickener, stabilizer, emulsifier, and texturizer in a variety of processed foods (including infant formula, whipped cream, cottage cheese, ice cream, and nutritional supplements) and other products (i.e., pharmaceuticals, cosmetics, and toothpaste) [
1‐
3]. Several international agencies have earlier reported that carrageenan is safe for human consumption [
4,
5]. However, investigators later demonstrated that carrageenan induces gastrointestinal ulcerations and cancer in animal models [
2,
4]. Carrageenan has been widely used in experimental models of inflammation to assess the activity of anti-inflammatory drugs and to study mediators of inflammation [
6]. Therefore, the concept of safety of carrageenan has remained highly controversial.
Current research studies on carrageenan have mainly focused on its effects at the organismal and cellular levels. Animal models such as rats [
6], guinea pigs [
7], pig [
8], mouse [
9], and rhesus monkeys [
10] have demonstrated that carrageenan induces colorectal tumors and ulcerative colitis. However, several international agencies have concluded that carrageenan is harmless [
4,
5]. The US Food and Drug Administration (FDA) has declared food-grade carrageenan as generally recognized as safe (GRAS) for use as a food ingredient [
11]. Tobacman et al. conducted extensive research studies to evaluate the safety of carrageenan at the cellular level [
2,
12‐
19]. They observed that exposure of human intestinal epithelial cells to food-grade carrageenan triggered a distinct inflammatory response via the activation of Bcl10 in the NF-κB pathway [
20], as well as up regulation of interleukin-8 (IL-8) secretion [
16,
20], and identified that the Toll-like receptor 4 (TLR4) is the surface membrane receptor for carrageenan in human colonic epithelial cells. Bernard et al. observed that carrageenan induces tumor necrosis factor alpha (TNF-α) production in human monocytes [
21]. Although an upregulation of inflammatory cytokines in human cells occurs following food-grade carrageenan exposure, we found that the increase was actually low, only 2–3-fold higher than that observed in the blank control [
16,
20]. Is this difference in expression level enough to cause a strong inflammatory reaction in vivo? Although lipopolysaccharides, which are highly proinflammatory molecules, can increase the secretion of IL-8 and TNF-α in macrophages by a dozen- to a hundred-fold [
22], the inflammatory mechanism of carrageenan remains elusive.
In our previous studies, we observed that λ-carrageenan enhances the LPS-induced production of IL-8 at the cellular level [
23]. Moreover, when differentiated human colonic epithelial cells were co-cultured with macrophage cells, the epithelial monolayers were significantly and easily damaged by carrageenan. Based on these earlier findings, we hereby propose the hypothesis that carrageenan is a conditional inflammatory agent. When the intestinal tract is in an “unhealthy” state such as that during bacterial infection or acute inflammation, carrageenan can synergistically enhance the inflammatory response.
Therefore, in the present study, an oxazolone (OXA)-induced mild intestinal inflammation model was developed to investigate whether κ-carrageenan can enhance the inflammatory responses of the gut with acute inflammation. Furthermore, the mechanism underlying this inflammatory response was also investigated in relation to the TLR4 pathway.
Methods
Animals
Seven-week-old male and female BALB/c mice with weights ranging from 20-25 g were purchased (SCXK 2012–0001, Beijing Weitong Lihua Experimental Animal Technology Co., Ltd.), housed in the animal center of Ningbo University Medicine College (Ningbo, China), and used in the present study. Five same-sex mice were housed in a cage and maintained under a 12-h light/12-h dark cycle (08:00 AM lights on) with food and water provided ad libitum. Housing and experimental environments were temperature- and humidity-controlled (21 ± 2 °C and approximately 60 %, respectively). All mice were allowed to habituate to the housing environment for 3 days prior to gavage administration. All experimental procedures were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and approved by the Ethical Committee of Animal Use and Protection of Ningbo University Health Science Center.
Animal experimental procedure
Eighty mice were divided into eight groups and placed in clear plastic cages with a mesh top under specific pathogen-free conditions, with each experimental group consisting of four males and four females. The experimental groups were designed as follows: the blank group only received a saline solution; the OXA group was treated with 150 μL of 1 % OXA; the carrageenan groups were treated with different concentrations of carrageenan, namely, the low-dose group (1.7 mg/kg, LOW), medium-dose group (8.3 mg/kg, MED), and high-dose group (41.7 mg/kg, HIG); carrageenan + OXA groups received low, medial, and high doses of carrageenan following by the administration of OXA and marked as “LOW + OXA,” “MED + OXA,” and “HIG+ OXA.”
κ-Carrageenan (Sigma Chemical Co., St Louis, MO, USA) was dissolved in saline with vigorous stirring at 37 °C and administered at three different doses, 1.7 mg/kg, 8.3 mg/kg, and 41.7 mg/kg by gavage administration. These doses were decided according to the data of preliminary experiments based on related study [
18]. To assess the synergistic effect of carrageenan on OXA-induced inflammation, different doses of carrageenan were gavage administered for two weeks prior to OXA administration.
OXA-induced intestinal inflammation was established as previously described by Heller et al., with minor modifications [
24]. To presensitize mice, a 2 × 2 cm area of the abdominal skin was shaved, and 200 μL of OXA [a 3 % (w/v) solution in 100 % ethanol, Sigma-Aldrich, St. Louis, MO] was smeared 5 days before OXA was administered. Five days after presensitization (day 0), the mice were rechallenged intrarectally with 100 μL of 1 % OXA in 50 % ethanol or only 50 % ethanol (vehicle, Blank group) under general anesthesia with isoflurane (Baxter, Deerfield, IL, USA). Intrarectal injection was administered with a polyurethane umbilical catheter. To ensure distribution of the OXA throughout the entire colon, the mice were held head down in a vertical position for 60 s after injection.
The mice were then monitored for symptoms of diarrhea, loss of body weight, and death the next day (day 1). The mice underwent exsanguination by cardiac puncture and the collected blood was used to determine the CD4+CD25+ Treg ratio and cytokine levels by flow cytometry. The colon was also isolated; one portion was used in macroscopic evaluation, whereas the rest was fixed in paraformaldehyde (Merck, Darmstadt, Germany) or glutaraldehyde solution for immunohistochemical and scanning electron microscopy analysis.
Macroscopic evaluation
Macroscopic evaluation of colonic tissues used the following criteria: 0, normal; 1, erythema only; 2, erythema, slight edema, and small erosions; 3, two or more bleeding ulcers and/or inflammation and/or moderate adhesions; and 4, severe ulceration and/or stenosis with dilations and/or severe adhesions.
Histological analysis
The colonic tissues were dissected and washed with Hank's balanced salt solution (containing 10 μg/mL of gentamicin, 100 U/mL of penicillin, and 100 μg/mL of streptomycin). The tissues were then fixed in 10 % natural buffered formalin, embedded in paraffin, cut into tissue sections (5-μm-thick), and stained with hematoxylin and eosin (H&E).
Histological examination was evaluated as previously described [
25]. The degree of histologic damage and inflammation was graded in a blinded fashion by expert histologists. The following manifestations were included in the evaluation: the amount of inflammation (0, none; 1, mild; 2, moderate; 3, severe; and 4, accumulation of inflammatory cells in the gut lumen), distribution of lesions (0, none; 1, focal; 2, multifocal; 3, nearly diffuse; and 4, diffuse), depth of inflammation and layers involved (0, none; 1, mucosa only; 2, mucosa and submucosa; 3, limited transmural involvement; and 4, transmural), and nature of mucosal changes (0, none; 1, minimal degeneration; 2, more degeneration; and 3, more necrosis). The overall histologic score was the sum of the four manifestations (maximum score: 15).
Intestinal mucosal morphology
The processing of intestinal segments for scanning electron microscopy was conducted as described elsewhere [
26]. Briefly, one segment of the tissue sample was fixed with 2.5 % glutaraldehyde. The segments were washed twice in phosphate buffer (pH = 7.2) and then post-fixed in osmium tetroxide (1 % in phosphate buffer, pH = 7.2, 2 h). After a series of dehydration steps in alcohol from 30 % to 100 %, the segments were critical point-dried (CPD030; BAL-TEC; Balzers; Liechtenstein), mounted on Al stubs, sputter-coated with gold by using a high-resolution fine coater (SCD005; BAL-TEC; Balzers; Liechtenstein), and then examined under a Jeol JSM-6301 scanning microscope (JEOL Ltd., Musashino, Tokyo, Japan). Morphological evaluation of the intestinal mucosal and microvilli was then performed. Scanning electron microscopic images (×200 magnification) were analyzed to measure the histological changes on the epithelial cell surface.
Determination of CD4 + CD25 + Treg ratio by flow cytometry
A total of 2.5 μL of anti-mouse CD3-eFluor450, 0.625 μL of anti-mouse CD4-PE, 0.625 μL of anti-mouse CD25-APC (all from eBioscience, California, USA) were added to 100 μL of peripheral blood. Following incubation at room temperature in the dark for 15–30 min, 1 mL of erythrocyte lysis solution was added to the samples and incubated for 15–20 min. The cells were washed with 2 mL phosphate-buffered saline (PBS) and resuspended in 500 μL PBS. The cells were then analyzed using a Beckman Gallios flow cytometer (Beckman Coulter, Inc., California, USA).
Immunohistochemical analyses
For immunocytochemical analysis, the mouse colonic tissues were fixed with 4 % paraformaldehyde in PBS, embedded in paraffin, and cut into tissue sections (5-μm-thick). The tissues were blocked with 1 % bovine serum albumin (BSA, Nacalai tesque) for 60 min at room temperature. The primary antibodies used in the present study included anti-TLR4 antibody (1:200, ab22048, Abcam, Cambridge, UK), anti-NF-κB antibody (1:200, 8242, CST, Danvers, MA, USA), anti-FOXp3 antibody (1:200, ab54501, Abcam, Cambridge, UK), and anti-PCNA antibody (1:100, Santa Cruz Biotechnology, Dallas, TX, USA). Secondary antibodies included goat anti-rabbit IgG (PV-9001, ZSGB-BIO, Peking, China) and rabbit anti-mouse IgG (PV-9005, ZSGB-BIO, Peking, China). Negative controls substituted with non-immunized mouse IgG at the same concentrations as that for each primary antibody were employed. Images were captured using a Leitz Dialux 22 Microscope (Leica Microsystems, Wetzlar, Germany) equipped with a QIcam Fast 1394 camera (QImaging, Surrey, BC, Canada). The images were analyzed using the software, Image Pro Plus 6.0 (Media Cybernetics UK, Marlow, UK). The levels of expression of TLR4, NF-κB, FOXp3, and PCNA were determined by measuring the positively labeled areas in relation to the total area.
Analysis of cytokine levels
Serum samples were collected after centrifugation of blood and stored at −80 °C. The levels of proinflammatory cytokines, including interleukin IL-6, tumor necrosis factor (TNF), Th1 (IL-2, interferon IFN-γ), Th2 (IL-4, IL-10), and Th17 (IL-17) cytokines in each serum sample were determined by using the BD Cytometric Bead Array mouse Th1/Th2/Th17 7 cytokine kit (BD BioSciences, NJ, USA) according to the manufacturer’s instructions. Samples and standards were analyzed on a FACS Calibur flow cytometer (BD BioSciences, NJ, USA). The concentrations were assessed by using FCAP Array software.
RNA extraction and real-time quantitative reverse transcription-PCR (RT-qPCR) analysis
Total RNA was isolated from scraped colonic mucosa of experimental mice using the TaKaRa RNAiso Plus Reagent (TaKaRa, Dalian, China) following the manufacturer’s instructions and then treated with RNase-free DNase I. Then, 2 μg of total RNA was re-transcribed into cDNA using a total reaction volume of 40 μL following standard M-MLV reverse transcriptase protocols (TaKaRa, Dalian, China). The primers of mouse target genes, including
TLR4,
NF-κB,
FOXp3, and
PCNA, as well as reference genes
β-actin,
GAPDH, and
18S-rRNA are listed in Table
1. The corresponding PCR products were sequenced by an ABI 3730 automated sequencer (Invitrogen, Carlsbad, CA, USA). To assess PCR efficiency, 10-fold serial dilutions of
TLR4,
NF-κB,
FOXp3,
PCNA,
β-actin,
Gapdh, and
18S-rRNA plasmid cDNA were used to generate a standard curve for each assay plate. The PCR reaction system included 1 μL of cDNA, 0.4 μM of forward and reverse primers, 10 μL of SYBR Premic Ex Taq II, 7.4 μL of dH
2O, as recommended by the manufacturer of SYBR-Green I (TaKaRa, Dalian, China). The cycling conditions were 95 °C for 5 min, followed by 40 cycles of 95 °C for 30 s, 60 °C for 1 min, and 72 °C for 15 s, which were conducted on a Mastercycler ep realplex real-time PCR system (Eppendorf, Hamburg, Germany). By using a standard curve, PCR efficiency was calculated (Table
1). After the amplification, melting curves were obtained by slow heating from 60 °C to 95 °C at increments of 0.5 °C/s and continuous fluorescence collection, which confirmed that only our specific product peaks were detected. RT-qPCR analysis of the samples was conducted as earlier described. Relative gene expression was analyzed by using the comparative cycle (Ct) value, which was compared using the relative quantification method. The mRNA expression of
TLR4,
NF-κB,
FOXp3,
PCNA,
GAPDH, and
18S-rRNA were normalized against the expression of
β-actin.
Table 1
Oligonucleotide primers used in this work
TLR4-F | GCTGCAACTGATGTTCCTTCT | 95 % | N/A | N/A |
TLR4-R | CCCAACATTCATCCATCTCA |
NF-κB-F | AAAGCCCTGACAGTCCATTG | 93 % | N/A | N/A |
NF-κB-R | TTGCTAGACACCGTCTGTGC |
FOXp3-F | TCGAGCTTCCCAGAGAGAGA | 94 % | N/A | N/A |
FOXp3-R | GGCCCTGACTGGATGTAAGT |
PCNA-F | AAGAAGAGGAGGCGGTAA | 92 % | N/A | N/A |
PCNA-R | AGTGTCCCATGTCAGCAA |
β-actin-F | TTGCTGACAGGATGCAGAAG | 94 % | N/A | N/A |
β-actin-R | ACATCTGCTGGAAGGTGGAC |
GAPDH-F | AATGTGTCCGTCGTGGATCT | 93 % | N/A | N/A |
GAPDH-R | GGTCCTCAGTGTAGCCCAAG |
18S-rRNA-F | ATTGGAGCTGGAATTACCGC | 95 % | N/A | N/A |
18S-rRNA-R | CGGCTACCACATCCAAGGAA |
To prepare a whole lysate solution, the collected tissues were washed twice with ice-cold PBS and lysed in a lysis buffer (Beyotime, Shanghai, China) that contained 1 mM of phenylmethylsulfonyl fluoride (PMSF) and a protein phosphatase inhibitor cocktail (Roche, Basel, Switzerland) for 10 min on ice. The homogenates were centrifuged at 12,000 rpm at 4 °C for 10 min. Total protein was extracted from the scraped colonic mucosa of the experimental mice using a NE-PER nuclear and Cytoplasmic Extraction kit (Pierce, Rockford, IL, USA), following the manufacturer's recommendations. Protein concentration was determined using a Bio-Rad DC Protein assay reagent (Bio-Rad Laboratories, Inc., Hercules, CA, USA), following the manufacturer's instructions.
Western blot analysis
Proteins (30 μg) in the nuclear extracts or in the whole lysate were separated by 10 % sodium dodecyl sulfate-polyacrylamide gel (SDS-PAGE), transferred onto polyvinylidene fluoride (PVDF) membranes, and probed with antibodies against TLR4, NF-κB, and p65 antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA), followed by appropriate horseradish peroxidase (HRP)-linked secondary antibodies (CST). The immunoreactive proteins were detected by using WesternBright ECL (Advansta Inc., Menlo Park, CA, USA). The results were quantified by measuring the band intensity relative to that of β-actin using the AlphaView™ Software (Alpha Innotech, San Leandro, CA, USA), and expressing as the relative intensities.
Statistical analysis
Statistical analyses were performed using the SPSS software, version 16.0 (SPSS Inc., Chicago, IL, USA). The results were expressed as means ± SEM. P < 0.05 was considered to indicate statistically significant differences. Macroscopic and histological score analyses were performed by using the student’s t-test. Differences in weight loss, cytokine levels, ratio of Treg (CD4+CD25+CD127dim), mRNA expression, western blot data, and average integral optical density of expression of various inflammatory proteins in mouse colonic mucosa among different groups were analyzed by using one-way ANOVA after post hoc testing.
Discussion
Although the FDA and JECFA consider carrageenan as a safe saccharide [
1,
5,
11], studies have shown that it induces ulcerations in animal models, suggests that it poses a carcinogenic risk to humans [
1‐
3,
27]. This finding has sparked controversy and intense debate, and the mechanism underlying this pathological condition has remained uncertain. The present study presents a possible mechanism of carrageenan that involves an inflammatory response.
BALB/c is a mouse strain that is susceptible to OXA-induced inflammation [
24,
28]. In this research, we used OXA to establish an acute gut inflammation BALB/c mouse model, which resembles ulcerative colitis in pathology and the immune response inof human. It has been reported that the initial toxic effects of OXA leads to a flooding of the lamina propria and inducing an increase in the secretion of Th2 cytokines IL-4 and IL-5, which in turn leads to further inflammation. [
24,
29]. Moreover, rectal instillation of 150 μL of 1 % OXA is enough to lead to an accelerated inflammation lasting for 1–2 days by a Th2 response in the distal half of the mouse colon of the mice, thereby resulting in either rapid recovery or death [
24]. Therefore, one day after injection of OXA was decided as the experimental time to detect the aggravating effect of carrageenan on the inflammatory response.
The average daily intake of carrageenan in the infant formula is 0.3 g/L [
2], amounting to 240 mg/5.8 kg/day (≈41.7 mg/kg/day). In the typical Western diet, the daily intake of carrageenan is considerably more than 500 mg/60 kg/day (≈8.3 mg/kg/day) [
30]. Bhattacharyya et al. reported that a dose of 50 μg/30 g/day (≈1.7 mg/kg/day) carrageenan induces colonic inflammation in IL-10-deficient mice [
18]. Therefore, in the present study, we used these concentrations to investigate the effect of carrageenan on OXA-induced inflammation.
Here, we first discovered that κ-carrageenan aggravates OXA-induced inflammation in BALB/c mice by significantly increasing weight loss, lethality rate, and degree of colonic injury. Of particular interest is our finding that mice pretreated with a medial-dose (8.3 mg/kg) of κ-carrageenan for 14 days resulted in a 37.5 % of mortality rate, whereas that of OXA-treated control mice was 12.5 % (Fig.
2). Histological and microscopic evaluation also confirmed that intragastric administration of κ-carrageenan aggravated OXA-induced inflammation in BABL/c mice. Of the three doses of κ-carrageenan used in the present study, the medium dose showed highest score histological scores. The survival ratio of the OXA + MED group was also the lowest. It is possible that the high concentration of carrageenan induces its formation into a gel in the gut of mice, which in turn reduces its contact without other gut content. In addition, Weiner et al. also observed that intragastric administration of a high dose of food-grade carrageen an just directly affected theresulted in the production of soft stools or development of diarrhea, which are common effects of non-digestible high-molecular weight compounds [
31]. Therefore, the aggravated effects of carrageenan do not appear to respond to a traditional dose–response curve.
OXA-induced inflammation is often employed in research investigations on the contribution of the Th2-dependent immune response to intestinal inflammation [
24]. The present study showed that OXA-induced inflammation is accompanied by the release of various inflammatory cytokines, not only Th2 type cytokines IL-10, but also pro-inflammatory cytokines, TNF-α and IL-6 [
24,
28]. Previous reports have revealed that OXA-induced inflammation is characterized by an upregulation of IL-4 [
29]; however, in the present study, although an increased level of IL-4 expression was observed in the OXA group, this was not statistically significant. No increase in the levels of Th1 cytokines IFN-γ, IL-2, and IL-17 was observed with OXA and κ-carrageenan treatment. The synergistic effect of κ-carrageenan on OXA-induced changes is also reflected in the secretion levels of inflammatory cytokines. The secretion of pro-inflammatory cytokines TNF-α and IL-6 in OXA-treated mice were all significantly augmented by κ-carrageenan pretreatment. These findings suggest that carrageenan aggravated the OXA-induced inflammation of intestinal tissues.
CD4
+CD25
+ regulatory T cells (Tregs) can regulate and suppress the immune response, thus; therefore, these play an important role in the maintenance of peripheral immune tolerance. The decrease in the number of Tregs often results in a development of various kinds of autoimmune disease such as inflammatory bowel disease (IBD) [
32]. The present study showed that the number of Tregs in the peripheral blood of OXA-treated group was lower than that of the Blank group, which further decreased after κ-carrageenan treatment, demonstrating that OXA can cause an imbalance in the immune system of mice, and carrageenan augments this imbalance.
FOXp3 is a transcription factor that plays a major role in the development and maintenance of the immunosuppressive function of Tregs [
33]. Galitovskiy et al. observed that the expression of FOXp3 is decreased in the inflamed areas of ulcerative colitis compared to that of the normal colon [
34]. This finding is consistentin agreement with the observations of the present study, wherein the intragastric administration of κ-carrageenan significantly decreased the OXA-induced mRNA expression of FOXp3 in the colonic mucosa, indicating that the suppressor activity of FOXp3 might have been abrogated in vivo or was insufficient to counterbalance the acute mucosal inflammation in OXA-treated mice.
Mice pretreated with different doses of carrageenan showed a significant increase in the levels of OXA-induced expression of NF-κB and TLR4. Some studies have demonstrated that carrageenan upregulated interleukin-8 (IL-8) secretion through TLR4, and activated NF-κB in human colonic epithelial cells [
16,
20]. Benard et al. also observed that κ-carrageenan induced TNF production via NF-κB activation [
21]. Recently, we have beenare exploring the mechanism of underlying the synergistic effect of κ-carrageenan on lipopolysaccharide (LPS)-induced interleukin-8 (IL-8) secretion in HT-29 cells. We found that κ-carrageenan pretreatment increased LPS-stimulated IL-8 secretion. The TLR4-NF-κB pathway was activated, as evidenced by the enhancement of Bcl10 (B-cell CLL/lymphoma 10) expression, IκBα phosphorylation, and nuclear NF-κB localization (Wu, wuweixiehou@163.com). These findings demonstrate that intragastric administration with κ-carrageenan significantly activated intestinal inflammation by upregulating TLR4 and NF-κB in colonic mucosa.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
WW, FW, and XG performed the experiments; TN and XZ analyzed the data; HC conceived and designed the research; WW and HC drafted the manuscript; and HC and XY edited and revised the manuscript. All authors read and approved the final manuscript.